S4.1 Social inequalities in extending working lives: the role of health. The EXTEND project
Chair: Dorly Deeg
Discussant: Sari Stenholm
The expected rise in pension costs related to population ageing has in many countries led to policy measures to extend working lives, including discouraging early retirement and raising statutory retirement age. The issue whether the level of health in the pertinent age group actually allows working longer is practically ignored. Health impairments are distributed unequally across socio-economic positions, and likewise, health impairments differ by gender. Hence, it is likely that policy measures that ignore social inequality, affect some workers’ ability to attain statutory retirement age and their quality of life before and after their exit from the workforce.
This symposium presents work from "EXTEND: Social inequalities in extending working lives of an ageing workforce". It addresses social inequalities in the potential and the conditions for extending working lives in five countries, with a central role for health and work conditions. Four studies are based on high-quality longitudinal datasets, and one study is a field experiment. Together, they guide directions for fairer measures to extend working lives.
S4.1.1 Explaining the social gradient in involuntary and voluntary retirement timing in
Jeevitha Yogachandiran Qvist
Aalborg University, Denmark
Background: A growing literature finds a social gradient in the timing of retirement. Yet, little previous work has examined what mechanisms explain this social gradient. This study identifies a number of mechanisms through which social class exerts its effect on involuntary and voluntary retirement. Methods: The study combines the Danish longitudinal Survey of Aging with register data. Event history analysis in a competing risk framework is applied with a sample of 3,340 older workers. Results: Workers in the working class retire involuntarily earlier than workers in the higher service class. There is also a social gradient in voluntary retirement, but the social gradient is less pronounced compared to involuntary retirement. The social gradient in involuntary retirement is partially explained by differences in physical and mental health, degree of volatility in employment careers, physical demands at work, and job control. The social gradient in retirement for voluntary reasons is explained by differences in job control. Conclusion: The results corroborate the existence of a social gradient in involuntary and voluntary retirement. The mechanisms found help to design measures to extend working lives.
S4.1.2 Drivers of involuntary and voluntary retirement for men and women in England
University of Sheffield, UK
Background: Previous research suggests that 1) health conditions are key drivers of retirement 2) the distinction between involuntary and voluntary retirement risk factors is important 3) men and women have distinct risk factors. This study therefore examines gendered risk factors for involuntary and voluntary retirement. Method: Competing risks survival analysis was used to test socio-demographic, health behaviour and health predictors of voluntary and involuntary retirement, separately for men and women. The analysis is based on English Longitudinal Study of Ageing data from 2002 with 12-year follow-up. Results: Regardless of gender, heart problems and depression increase the risk of involuntary retirement, and having a retired partner, caring responsibilities and high SES increase the risk of voluntary retirement. Diabetes showed twice the risk of involuntary retirement for women. Health behaviours had little influence. The population attributable fraction of these risk factors varies widely given strong gender differences in their population rates: women have double the rate of depression and care responsibilities, but half the rate of diabetes. Conclusions: This study confirms the importance of distinguishing voluntariness and gender in retirement timing. The results suggest ways to design policy and intervention strategies to reduce early retirement.
S4.1.3 Work stress among older employees in Germany: Social inequalities in health and
University of Cologne, Germany
Background: Previous research has shown that a main reason for early retirement is poor health. Health in turn is influenced by exposure to the job environment. Furthermore, both factors vary by socioeconomic background. This study examines the relationship between work stress and retirement age, and whether this relationship is mediated by health. Methods: A German subsample of the Survey of Health, Aging and Retirement in Europe is linked with register data of the German Public Pension Scheme. The sample includes 304 individuals initially aged 50-65, followed from 2004 to 2015. Structural equation modelling is applied to analyse the direct and indirect effects of SES and work stress on retirement age via health. Results: Only one dimension of job stress, namely higher job strain, relates to poorer health and lower retirement age. Effort-reward imbalance has no effect. Health does not operate as mediator in the relationship between work stress and retirement age. Conclusions: Lowering job strain potentially extends people’s work life, as it made a direct contribution in explaining early retirement age.
S4.1.4. How do work characteristics affect post- work exit health in low and high SES
Sascha de Breij1, Dan Holman2, Jeevitha Yogachandiran Qvist3, Dorly Deeg1
1 VU University, Amsterdam, The Netherlands, 2 University of Sheffield, UK, 3 Aalborg University, Denmark
Background: Socioeconomic inequalities in health have been widely reported. A low socioeconomic position (SEP) is associated with more adverse working conditions. Working conditions are associated with health. This studied examined (1) socioeconomic differences over time in health after work exit and (2) to what extent work characteristics mediate these SEP inequalities in health. Methods: Three longitudinal datasets were used: The Netherlands (LASA), Denmark (DLSA) and England (ELSA). We used Generalized Estimating Equations (GEE) to examine the relation between SEP and self-rated health (SRH) and functional limitations after work exit (with maximum follow-up of 18 years) and possible mediation of work characteristics (physical demands, psychosocial demands, autonomy, variation in activities and type of work exit). Results: Low SEP participants reported significantly poorer SRH and more functional limitations than high SEP participants. Physical demands, autonomy, variation in activities and type of work exit (disability, unemployment, early retirement, regular retirement) were found to be mediators in the relation between SEP and health after work exit. Conclusion: Social inequalities in health are still present after work exit. Work characteristics act as mediators and should therefore receive more attention to reduce these health inequalities.
S4.1.5 Efficacy of a participatory training intervention on well-being and career
indicators among senior employees
Mervi Ruokolainen, Jukka Vuori
Finnish Institute of Occupational Health, Finland
Background: This study examines the effects of group training techniques of the “Engagement for Late Career” (ELC) program on senior employees’ (55+) well-being and career-related factors. The ELC aims to enhance employees’ personal resources in a participatory group process at workplaces, by motivating and supporting senior employees in retaining their working career until and beyond the retirement age, and manage the challenges that threaten it in the current career environment. It has previously been shown that employees with less resources benefit from the training most. Methods: Data were collected in a randomly assigned field experimental study between 2014 and 2016 among 329 employees in 31 training groups in 17 Finnish organizations. The participants evaluated the training (i.e., active learning methods, trainer’s skill, preparation for setbacks, supportive learning environment, and skill training) after the last session. Well-being outcomes were depression and perceived workability and career-related outcomes, future time perspective at work and retirement thoughts. Results: Preliminary results show that the participants’ positive perceptions of the skill training and preparation for setbacks predicted better workability and longer future time perspective at work at six-months follow-up. Next, effects will be analysed at group level using multilevel modelling. Conclusions: ELC appears a recommendable resource-based peer-group program for enhancing senior employees’ well-being and longer working careers.